This is why Hayek called competition "a discovery procedure" in which the number of firms are irrelevant. Unfortunately, since the 1920s most economists have — thanks to Frank Knight — come to think of competition as a quantitative state rather than a process. George Stigler was one such economist — until he significantly altered his views on monopoly and mergers. He stated somewhat ruefully that

I now marvel at my confidence at that in discussing the proper way to run a steel company. I certainly would not presume today to have that knowledge about any industry, even higher education. What is still more embarrassing is that I no longer believe the economics I was preaching… (Memoirs of an Unregulated Economist, Basic Books Inc., 1988, P. 99).

Unfortunately for Australia the likes of Fels and Samuel lack Stigler's capacity for changing his mind once reality refutes theory. The quantitative approach assumes that unless there are a large number of producers in the market there is no true competition. It therefore reasons that a market with few firms will set prices and cause output to diverge from the competitive level. There is absolutely no justification for this belief. What matters is not numbers but contestability. As long as potential competitors are not hindered by state imposed obstacles from competing with the incumbents the market is still competitive — even if it is being served by only one producer.

(The fallacious reasoning behind the concept of perfect competition has been compounded by the theory of monopolistic competition).

Samuel's argument is the hoary one that cartels are organised to raise prices and restrict output. But the one question that never seems to get asked is how can a cartel raise prices above the market clearing level in a free market? If several companies combine to form a cartel then any attempt on their part to raise prices will only encourage others to enter the market place and drive prices back down again. Cartels in the anti-social sense can only emerge if competitors are kept out by force. And only the state can wield that kind of power. Even then problems will emerge.

Let us say a number of companies form, with the support of the state, a cartel. Raising prices forces the cartel to curb output. This means that the more efficient members of the cartel will have to curb output to a greater extent then the less efficient members if the cartel's numbers are to be maintained. But why should they? Eventually, the more efficient members would break away (sometimes after a great deal of cheating on their quotas) so that they can fully exploit their superior efficiency.

This is basically why cartels are inherently unstable. If the state intervenes to maintain the cartel, subsidies will have to be paid to compensate for lost output. The point, however, is that the cartel can only exist with the support of state power, as was the case in Bismarck's Germany. (Germany used exports to overcome the output problem).

This brings us back to our trucking companies. That they formed a cartel-like arrangement is indisputable. So what? So long as others were free to compete against them their actions cannot be seen as anti-competitive, except by those — like Fels and Samuel — who do not understand the nature of competition.

What critics overlook is that forming a cartel can be a tactic to evade merger laws. In other words, what appears to be a cartel can actually be a substitute for a merger or a form of producers' cooperative, which is also a form of merger. In these cases assets are pooled so that they can be more efficiently allocated.

The work of Professor Coase can be of some help here. He showed that firms internalise certain market transactions and this, he believed, was the reason for the existence of the firm. (I think he went to far here. Even in the absence of transaction costs firms would still emerge). It takes but a little thought to realise that the kind of voluntary behaviour that Fels dragged before the court might also have had the effect of lowering the costs of production by eliminating many transaction costs. After all, as Coase pointed out, firms have a tendency to expand until the internal costs of doing so equal transaction costs. Now this also fits in with Coase's illuminating observation that government

planning is imposed on industry, while firms arise voluntarily because they represent a more efficient method of organising production. In a competitive system there is an 'optimum' amount of planning. (R. H. Coase The Firm, the Market and the Law, University of Chicago Press, 1990, p. 37)

But when we say "firms arise voluntarily" this must also include the kind of cooperative activity Samuel is attacking as anti-competitive! There is, in principle, no difference between several firms pooling their assets (fixed and circulating capital) than several people pooling their liquid assets (cash) to form a corporation. Both are pooling savings because fixed and circulating capital are savings. In other words, the actions of Samuel might very well be preventing firms from reaching their optimum size.

The conclusion is straightforward. A cartel cannot suppress competition. This is a fact that the likes of Samuel never address. Even if we accept the argument that a cartel means the absence of competition among its members this cannot logically be taken as meaning the absence of competition in the market place. To make this assumption Samuel would have to explain why cooperative action that generates profits does not attract competitors.

Gerard Jackson is Brookesnews' economics editor

Subscribe to BrookesNews Bulletin